metadata
base_model:
- Finnish-NLP/sft-hf_hf_2024_06_29_14_08_01_checkpoint_564_dpo-checkpoint-249
- Finnish-NLP/sft-hf_hf_2024_07_09_16_33_52_checkpoint_1758_dpo-checkpoint-832
tags:
- merge
- mergekit
- lazymergekit
- Finnish-NLP/sft-hf_hf_2024_06_29_14_08_01_checkpoint_564_dpo-checkpoint-249
- Finnish-NLP/sft-hf_hf_2024_07_09_16_33_52_checkpoint_1758_dpo-checkpoint-832
Ahma-3B-dpo-slerp
Ahma-3B-dpo-slerp is a merge of the following models using LazyMergekit:
- Finnish-NLP/sft-hf_hf_2024_06_29_14_08_01_checkpoint_564_dpo-checkpoint-249
- Finnish-NLP/sft-hf_hf_2024_07_09_16_33_52_checkpoint_1758_dpo-checkpoint-832
🧩 Configuration
slices:
- sources:
- model: Finnish-NLP/sft-hf_hf_2024_06_29_14_08_01_checkpoint_564_dpo-checkpoint-249
layer_range: [0, 26]
- model: Finnish-NLP/sft-hf_hf_2024_07_09_16_33_52_checkpoint_1758_dpo-checkpoint-832
layer_range: [0, 26]
merge_method: slerp
base_model: Finnish-NLP/sft-hf_hf_2024_06_29_14_08_01_checkpoint_564_dpo-checkpoint-249
parameters:
t:
- filter: self_attn
value: [0, 0.5, 0.3, 0.7, 1]
- filter: mlp
value: [1, 0.5, 0.7, 0.3, 0]
- value: 0.5
dtype: bfloat16
💻 Usage
!pip install -qU transformers accelerate
from transformers import AutoTokenizer
import transformers
import torch
model = "RASMUS/Ahma-3B-dpo-slerp"
messages = [{"role": "user", "content": "What is a large language model?"}]
tokenizer = AutoTokenizer.from_pretrained(model)
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
pipeline = transformers.pipeline(
"text-generation",
model=model,
torch_dtype=torch.float16,
device_map="auto",
)
outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])